Guide Labs debuts a new kind of interpretable LLM | TechCrunch
Summary
Guide Labs introduces Steerling-8B, an open-sourced interpretable LLM designed to enhance understanding of AI model outputs by tracing token origins back to training data.
Why It Matters
As AI models become more complex, understanding their decision-making processes is crucial for trust and safety. Guide Labs' Steerling-8B aims to address interpretability challenges, which is vital for regulated industries and applications requiring transparency.
Key Takeaways
- Steerling-8B is an 8 billion parameter LLM focused on interpretability.
- The model allows tracing outputs back to training data, enhancing transparency.
- Interpretability is essential for regulated industries like finance and healthcare.
- The architecture may retain emergent behaviors while improving control over outputs.
- Guide Labs' approach could set a new standard for AI model development.
The challenge of wrangling a deep learning model is often understanding why it does what it does: Whether it’s xAI’s repeated struggle sessions to fine-tune Grok’s odd politics, ChatGPT’s struggles with sycophancy, or run-of-the-mill hallucinations, plumbing through a neural network with billions of parameters isn’t easy. Guide Labs, a San Francisco start-up founded by CEO Julius Adebayo and chief science officer Aya Abdelsalam Ismail, is offering an answer to that problem today. On Monday, the company open-sourced an 8 billion parameter LLM, Steerling-8B, trained with a new architecture designed to make its actions easily interpretable: Every token produced by the model can be traced back to its origins in the LLM’s training data. That can as a simple as determining the reference materials for facts cited by the model, or as complex as understanding the model’s understanding of humor or gender. “If I have a trillion ways to encode gender, and I encode it in 1 billion of the 1 trillion things that I have, you have to make sure you find all those 1 billion things that I’ve encoded, and then you have to be able to reliably turn that on, turn them off,” Adebayo told TechCrunch. “You can do it with current models, but it’s very fragile … It’s sort of one of the holy grail questions.” Adebayo began this work while earning his PhD at MIT, co-authoring a widely cited 2018 paper that showed existing methods of understanding deep learning models were not reliable. That work ultimat...
